The invention relates to a method for distinguishing between a movement state and a rest state of a motor vehicle by evaluating the output signal of a sensor arranged in the vehicle.
Particularly in connection with navigation systems of motor vehicles, a plurality of sensors are used for determining the spatial position of a vehicle. The exact position of the vehicle is usually determined from a combination of satellite position-finding and dead reckoning position-finding.
For dead reckoning position-finding, use is made of a direction sensor, which is usually designed as a rate-of-rotation sensor, and a distance measuring device. The distance measuring device may be, for example, the tachometer which is present in any case in the vehicle. On account of the greater accuracy, however, the measurement signals of an anti-lock braking system are also often used for distance measurement. However, it is precisely in the case of the latter system that signals are suppressed below a specific limiting speed of typically about 3 km/h. This means that the vehicle standstill cannot be reliably identified using the signals of the distance measuring device.
Identification of the vehicle standstill is of particular importance, however, in navigation systems, since the various position-finding signals are calibrated in this state. Moreover, interference signals during the vehicle standstill, such as e.g., caused by the slamming of the vehicle door, leads to erroneous calibration. It is desirable, therefore, to carry out calibration of the sensors only when the motor vehicle is in a rest state, i.e., the vehicle is both at standstill and, preferably, no other vibrations occur.
The object of the invention, therefore, is to specify a method which makes it possible to reliably distinguish between a movement state and a rest state of a motor vehicle, this distinguishing being possible, in particular, even in the absence of a signal from a distance measuring device.
In the case of a method of the generic type, the object is achieved by virtue of the fact that the noise of the sensor signal is evaluated in order to distinguish between the movement state and the rest state. It has been found that the noise of a sensor in the motor vehicle distinctly depends on whether the motor vehicle is in a movement state or a rest state. Proceeding from this insight, therefore, the sensor noise is preferably evaluated statistically. This makes it possible to reliably distinguish between a movement state and a rest state. In particular, the rest state in this case does not just include the vehicle standstill, rather external vibrations of the vehicle as are produced e.g. ,by the slamming of a vehicle door can also be reliably detected using the statistically evaluated noise. Depending on the statistically evaluated sensor noise, a comparator outputs a yes/no signal with regard to the movement state of the motor vehicle.
The statistical evaluation of the sensor noise is effected, in particular, in such a way that firstly the mean is calculated from the signals Xi of the sensor in accordance with       m    x          (      1      )        =            x      _        =                  1        N            ⁢                        ∑                      i            =            1                    N                ⁢                  xe2x80x83                ⁢                  x          i                    
and the mean is subsequently squared. In addition, the root mean square is calculated from the sensor signals in accordance with       m    x          (      1      )        =            x      _        =                  1        N            ⁢                        ∑                      i            =            1                    N                ⁢                  xe2x80x83                ⁢                  x          i                    
Finally, the variance is calculated from the squared mean and the root mean square in accordance with       σ    x    2    =            (              m        x                  (          2          )                    )        -                            (                      m            x                          (              1              )                                )                2            .      
The variance thus formed is fed to a comparator which, by comparison with a stipulated value, takes a yes/no decision with regard to the movement state.
The sensor signal may, in particular, be the signal of a rate-of-rotation sensor, as is used in navigation systems. In this case, the identification of the rest state of the motor vehicle is performed entirely without any consideration of the signals from a distance measuring device. In particular, the method can, therefore, also be used when, in the navigation system, instead of a distance measuring device, the signals from an acceleration sensor are evaluated for the distance determination.
The method is preferably used in a navigation system having a plurality of sensors for position determination. These sensors are, in particular, a direction sensor and a distance and/or acceleration sensor. A distinctly improved accuracy can be achieved by calibrating these sensors in the rest state of the vehicle whilst taking account of the position signals of a satellite navigation system. The sensors of the navigation system are calibrated in a manner known per se. In order to achieve high accuracy in the determination of the position of the vehicle, dead reckoning navigation and satellite navigation are additionally accompanied by so-called map matching, in which a comparison of the vehicle position determined by dead reckoning and/or satellite navigation with road map data is carried out in order to determine the vehicle position.
However, the application of the method is not restricted to navigation systems. As a result of reliably distinguishing between a movement state and a rest state of the vehicle, the method can also be utilized for the purpose of detecting vehicle vibrations in the case of a vehicle which is not being operated. In particular, the method can, therefore, be used in an alarm system which, with the vehicle parked, detects vehicle vibrations and, if appropriate, triggers an alarm. The alarm triggering threshold can be set by corresponding setting of the comparator, so that slight vehicle vibrations, as occur e.g. as a result of trucks traveling past, do not lead to alarm triggering.
A navigation system, according to the invention, for motor vehicles having sensors for determining a vehicle position is characterized in that the navigation system contains means for evaluating the noise of a sensor signal and for carrying out standstill identification using the evaluated sensor noise signal. In particular, means for calibrating the sensors are furthermore present in the navigation system. The corresponding means may be embodied either as software or else at least partly as hardware.